It has repeatedly been reported that snow cover is a dominating factor in determining the presence or absence of permafrost in the discontinuous and sporadic permafrost regions. The temperature at the snow-soil interface by the end of winter, known as the bottom temperature of winter snow (BTS) method, has been used to detect the existence of permafrost in European alpine regions when the maximum snow depth is about 1.0 m or greater. A critical snow thickness of about 50 cm or greater can prevent the development of permafrost in eastern Hudson Bay, Canada. The objective of this study is to investigate the impact of snow cover on the presence or absence of permafrost in cold regions through numerical simulations. A one-dimensional heat transfer model with phase change and a snow cover regime is used to simulate energy exchange between deep soils and the atmosphere. The model has been validated against the in situ data in the Arctic. The simulation results indicate that both snow depth and the onset date of snow cover establishment are important parameters in relation to the presence or absence of permafrost. Early establishment of snow cover can make permafrost disappear, even with a relatively thin snow cover. Permafrost may survive when snow cover starts after the middle of December even with a snow thickness >1.0 m. This effect of snow cover on the ground thermal regime can be explained with reference to the pattern of seasonal temperature variation. Early establishment of snow cover enhances the insulating impact over the entire cold season, thus warming and eventually thawing the permafrost. The insulating effect is substantially reduced when snow cover starts relatively late and snowmelt in the spring creates a huge heat sink, resulting in a favorable combination for permafrost existence.